[unreadable] [unreadable] The decision to surgically decompress the spinal cord following spinal cord injury (SCI) is controversial, as is the timing of decompression. Multicenter, randomized controlled studies of early surgical decompression have not been performed in humans. MR images of the spine reflect the anatomical picture of the spinal cord but do not reflect metabolic status of the tissue or the CSF surrounding the spinal cord and thus may not indicate the precise degree of compression on the spinal cord. We hypothesize that: [unreadable] 1) quantitative cine phase-contrast magnetic resonance imaging (C-MR), which non-invasively measures flow velocity of cerebrospinal fluid (CSF) above and below the injury, will provide additional objective data to assist the decision-making process for surgical decompression and whether use of this data results in better long-term functional outcome following SCL. Within 8 hours after admission, all patients will undergo T1- and T2-weighted MR scans followed immediately by a C-MR. Patients with cord compression and evidence of CSF block will undergo immediate decompression surgery. Patients with no MR evidence of spinal cord compression or CSF block will be treated with the best medical care (non-surgical). Patients in whom the anatomical picture (MR scan), and CSF flow pattern (C-MR) are not concordant will be randomized into surgical and non-surgical groups. A variety of functional outcome measures will be performed on admission, weekly for one month, and at 3, 6, 9, 12, 18, and 24 months following SCI, to determine whether use of C-MR data results in a better functional outcome. [unreadable] 2) integrated PET/CT can determine the metabolic status at the site of the injury (as early as possible, but within 3 days) after the injury. We will use 18F-deoxyglucose to assess uptake and utilization. This information will attempt to assess whether the injury is complete (irreversible) or incomplete (reversible). We will answer the question if PET/CT is performed early can it be used to predict the outcome of spinal decompression surgery? [unreadable] 3) tcMMEP and SSEP can be used as non-invasive tools to assess spinal compression. We will measure them on the day of admission and after decompression, and correlate them post hoc with the MR and C-MR data, all of which will be entered into an NIH-sponsored national SCI database. We will also assess whether these tests can be used to more accurately determine if decompressive surgery can improve the long-term functional outcome by performing these tests weekly for one month, and at 3, 6, 9, and 12 months following SCI. [unreadable] 4) oximetry at the site of SCI can be used over the first week as a non-invasive monitor of cord ischemia that results from compression and, thus, as a decision-making tool for decompression. This can be done by correlating the data of each patient [unreadable] post hoc with other measures of Aims 1-3. [unreadable] [unreadable]